Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Unclear on the limitations of Filo
#1
So in the docs it says that Filo cannot be used for free cable simulation. And then I've seen you post that the requirement is that you have to have objects connected to both ends of the rope. It seems to me then that you could have just two free (dynamic) rigid bodies with a cable linking them, is that possible? It seems like maybe not because in all of your examples, there's at least one static object. Is this a requirement or was that just happen stance? 

You also linked a paper in one of these threads that demonstrated a bow made from cables where it appeared as though the bow was a dynamic object. Would it be possible to make something like this possible with Filo? 

Thanks so much for any illumination!
Reply
#2
(14-08-2021, 04:43 AM)scarlet Wrote: So in the docs it says that Filo cannot be used for free cable simulation. And then I've seen you post that the requirement is that you have to have objects connected to both ends of the rope. It seems to me then that you could have just two free (dynamic) rigid bodies with a cable linking them, is that possible?

By "free" cables, I mean cables that are attached to nothing, hence simulated by themselves. You can attach Filo cables to both dynamic and static objects. Having two dynamic objects attached to the ends a cable is perfectly possible.

The main difference between Filo and other simulators is that Filo does not actually simulate the cable itself. It simulates the effect a cable would have on a list of objects in contact with or attached to the cable. That's why, if there are no objects attached to the cable, there's no simulation to perform. This is also the reason why the cost of the simulation is independent of cable length, and only depends on the amount of objects in contact with the cable.

(14-08-2021, 04:43 AM)scarlet Wrote: It seems like maybe not because in all of your examples, there's at least one static object. Is this a requirement or was that just happen stance? 
Just a coincidence, there' no reason that would prevent all objects from being dynamic.

(14-08-2021, 04:43 AM)scarlet Wrote: You also linked a paper in one of these threads that demonstrated a bow made from cables where it appeared as though the bow was a dynamic object. Would it be possible to make something like this possible with Filo? 

Yes, Filo is a direct implementation of the method described in that paper so everything shown in it can be simulated.

let me know if I can be of further help!
Reply
#3
(16-08-2021, 08:24 AM)josemendez Wrote: By "free" cables, I mean cables that are attached to nothing, hence simulated by themselves. You can attach Filo cables to both dynamic and static objects. Having two dynamic objects attached to the ends a cable is perfectly possible.

The main difference between Filo and other simulators is that Filo does not actually simulate the cable itself. It simulates the effect a cable would have on a list of objects in contact with or attached to the cable. That's why, if there are no objects attached to the cable, there's no simulation to perform. This is also the reason why the cost of the simulation is independent of cable length, and only depends on the amount of objects in contact with the cable.

Just a coincidence, there' no reason that would prevent all objects from being dynamic.


Yes, Filo is a direct implementation of the method described in that paper so everything shown in it can be simulated.

let me know if I can be of further help!
Awesome, thanks for the insight! That's really, really cool.
Reply